Introduction

Brown algae (Phaeophyceae) have successfully occupied almost all marine coasts in cold, temperate and tropical zones, and play important ecological roles in the marine ecosystem. They have developed diverse morphology from the giant kelp of the order Laminariales (Macrocystis and Nereocystis) to the microscopic, filamentous representatives of the Ectocarpales (Laminariocolax and Microsyphar) (Charrier et al. 2012). So far, brown algae comprise a multitude of taxa including 19 orders, 62 families, 473 genera and nearly 2000 species (Guiry and Guiry 2014). Phylogenetic analyses are required to develop a complete inventory of the Phaeophyceae to enable us to understand evolutionary biodiversity over time (Silberfeld et al. 2014). Many attractive features of the mitochondrial genome (e.g., its abundance in the cell, accelerated substitution rates, and genome-level information content) have made it an invaluable tool in phylogenetics, phylogeography, and population genetics at various taxonomic levels (Liu et al. 2015; Liu and Pang 2015).

Complete mtDNA sequences from under-investigated categories would supplement and improve our understanding of mitochondrial genome evolution in the Phaeophyceae. Its remarkable stability and universality of mitogenome content in major groups of stramenopiles have allowed analyses of mitochondrial genome sequences and gene order to reconstruct deep node phylogenies and to reassess relationships among closely related species (Ravin et al. 2010; Liu et al. 2015). However, there is still scarce information on mitogenomes of brown algae, and only 25 complete brown algal mtDNAs have been reported (Oudot-Le Secq et al. 2001; Yotsukura et al. 2010; Zhang et al. 2013; Liu and Pang 2015). The known brown algal mitochondrial genomes are circular, double-stranded DNA molecules, ranging in length from 31.6 to 58.5 kb (Table 1). Their organization and gene content are highly conserved. Brown algal mtDNAs usually harbor three ribosomal RNA genes, 24–26 transfer RNA genes, 35–36 protein-encoding genes, several open reading frames (ORFs), and some non-coding regions containing signals for transcription and replication.

Table 1 The 20 brown algal species with complete mitochondrial genomes comparatively analyzed in this study
Full size table

Turbinaria J.V. Lamouroux is a relatively species-poor genus of the order Fucales (Wynne 2002; Le Lann et al. 2008), and to date, there are 22 species accepted taxonomically in Turbinaria (Guiry and Guiry 2014). Turbinaria species display great morphological plasticity in response to different environmental conditions, thus molecular data are used to delineate species boundaries (Prathep et al. 2007; Rohfritsch et al. 2007, 2010; Draisma et al. 2010). Turbinaria ornata (Turner) J. Agardh is a perennial brown alga native to coral reef ecosystems of tropical areas of the Pacific and Indian Ocean. Since the 1980s, this alga has markedly expanded its abundance and geographical distribution in French Polynesia, due to dispersal of detached, reproductively mature floating fronds (Stewart 2004, 2006; Stiger and Payri 2005). The aggregations of this alga could provide the spatial refuge from herbivory to other algal species on coral reefs and increasing local algal species richness (Bittick et al. 2010). Despite its ecological importance, very little is known about its genome sequence, which has restricted our understanding of its genetic diversity, interspecific taxonomic positions and evolutionary relationships.

In this study, the complete mitochondrial genome sequence of T. ornata has been determined, and comparatively analyzed with published brown algal mtDNAs from Fucales, Desmarestiales, Ectocarpales, Laminariales and Dictyotales, to gain more general insights into the evolution of brown algal mitogenomes. To our knowledge, this is the first report of a completely annotated mitogenome sequence for the genus Turbinaria.

Materials and methods

Sample collection and DNA extraction

Adult plants of Turbinaria ornata (Turner) J. Agardh were collected on the intertidal rocks from Dadonghai Bay, Sanya, Hainan Province, China (37°09′N, 122°32′E) on 1st April, 2014. Plants were transported to the laboratory in coolers (5–8 °C) within 48 h after collection. Fresh tissue was ground to a fine powder in liquid nitrogen for DNA extraction. Total genomic DNA was extracted using a Plant Genomic DNA Kit according to the manufacturer’s instructions (Tiangen Biotech, Beijing, China). The concentration and quality of isolated DNA were evaluated via electrophoresis on a 1.0 % agarose gel.

PCR amplification and sequencing

Based on the known mitogenome sequence data in the Fucales, seven pairs of PCR primers were designed to amplify the entire T. ornata mitogenome with lengths of 7.3, 7.6, 4.0, 1.9, 4.9, 5.9 and 6.8 kb. The primer pairs were as follows: T-23S-F (5′-CTTGGAAGCAGCCATTCTT-3′, this study)/T-trnQ-R (5′-ACCAAAAACGGGCGCCTTAC-3′, this study), T-trnW-F (5′-AACTTTGGTTTCCAAAGCCAACG-3′, this study)/T-atp6-R (5′-ATAAGTCAAATAAACCAATACCT-3′, this study), CAF4A (Kogame et al. 2005)/cox1-R (5′-AACCAAAACAAATGCTGATAAA-3′, this study), T-cox1-F (5′-GGCATCACTATGCTTTTAACGG-3′, this study)/T-cob-R (5′-CAATAATATGAGAATCTACCCAAC-3′, this study), T-nad9-F (5′-CATTTGAGGTGTTTGGACATGAG-3′, this study)/T-nad4-R (5′-CCAAAAACACCAATAATTAGAAGC-3′, this study), cox2-F/trnS-R (Liu et al. 2015), and T-trnS-F (5′-GGAGATGTGGCTGAGTGGCT-3′, this study)/T-23S-R (5′-CTCTTCTCGATGTAGCTACCCG-3′, this study).

PCR was performed in 50 μL reactions containing 32 μL of sterile distilled H2O, 10 μL of 5× PrimeSTAR GXL buffer (5 mM Mg2+ plus, Takara, Japan), 4 μL of dNTP mixture (2.5 mM each), 1 μL of each primer (10 μM), 1 μL of PrimeSTAR GXL DNA polymerase (1.25 units μL−1, Takara, Japan), and 1 μL of DNA template (approximate 50 ng). PCR amplification was carried out on a T-Gradient Thermoblock Thermal Cycler (Whatman Biometra, Goettingen, Germany) with an initial denaturation at 94 °C for 3 min, followed by 30 cycles of denaturation at 94 °C for 20 s, annealing at 50–52 °C for 50 s, extension at 68 °C for 1 min/kb, and a final extension at 68 °C for 10 min. PCR products were purified using the Qiaquick Gel Extraction Kit (Qiagen, Germany). Sequencing reactions were performed using ABI 3730 XL automated sequencers (Applied Biosystems, USA).

Genome assembly and annotation

The DNA sequences were manually edited and assembled using the BioEdit program (http://www.mbio.ncsu.edu/BioEdit/bioedit.html). DNA sequences of the complete mitochondrial genome of T. ornata were determined by comparison with published mitochondrial genome sequences of the Fucales (Oudot-Le Secq et al. 2006; Liu et al. 2014a, c, 2015; Liu and Pang 2015). The locations of the protein-coding genes were determined with DOGMA (Wyman et al. 2004), ORF Finder, and Blastx searches (GenBank). Transfer RNA genes were identified by reconstructing their cloverleaf structures using the tRNAscan-SE 1.21 software with default parameters (Schattner et al. 2005). Ribosomal RNA genes were identified by BLAST searches (Altschul et al. 1997) of the nonredundant databases at the National Center for Biotechnology Information. The map of the mitochondrial genome was built using Organellar Genome DRAW (Lohse et al. 2013). The complete mitogenome sequence of T. ornata is deposited in GenBank with the accession number KM501562.

Phylogenetic analyses

Phylogenetic analyses were carried out using 20 reported brown algal mitogenome sequences (Table 1). Sequences of individual genes were separately aligned and manually adjusted using MEGA 5.2 software (Tamura et al. 2011). The sequences of 21 brown algal mitochondrial genomes were compared using Blast Ring Image Generator (BRIG) software (Alikhan et al. 2011). Phylogenetic relationships within Phaeophyceae were analyzed based on four datasets: (1) three ribosomal RNA genes (rnl, rns, and rrn5), (2) 25 conserved transfer RNA genes, (3) 35 functionally known protein-coding genes and (4) their concatenated protein sequences (rps2-4, 7, 8, 10-14, 19; rpl2, 5, 6, 14, 16, 31; nad1-7, 9, 11; cob; cox1-3; atp6, 8, 9; and tatC). Each dataset was subjected to concatenated alignments using ClustalX 1.83 with the default settings (Thompson et al. 1997). Base composition and pairwise comparison were examined by MEGA 5.2. The identity percentages of 21 brown algal mtDNA sequences including 3 rRNA, 25 tRNA and 35 protein-coding genes were evaluated using the BioEdit v7.0.8.0 software (Hall 1999). Neighbor-joining (NJ) and maximum likelihood (ML) trees were conducted with 1000 bootstrap replicates using MEGA5.2. Dictyota dichotoma served as the outgroup according to Oudot-Le Secq et al. (2006). The evolutionary distances were computed based on the Kimura two-parameter model (Kimura 1980) for DNA sequences and on the Whelan And Goldman + Freq. model (Whelan and Goldman 2001) for protein sequences. Final results are presented using gaps as fifth character states for bases in DNA sequences.

Results and discussion

Genome features

The circular T. ornata mitogenome is 34,981 bp in length within the range of known Fucales mitogenomes from 34,609 bp in Sargassum horneri to 36,392 bp in Fucus vesiculosus. The reported Fucales mtDNAs are shorter than that of Desmarestiales, Ectocarpales (not including Colpomenia peregrina), and Laminariales (Table 1). The overall AT content of the T. ornata mitogenome is 64.18 % and falls between Sargassum (62.47–63.84 %) and F. vesiculosus (65.55 %). The T. ornata mitogenome harbors 65 densely packed genes which are the same genes found in species of Sargassum (Liu et al. 2015; Liu and Pang 2015), including 35 protein-coding genes, three rRNAs (rnl, rns and rrns), 25 tRNAs, and two conserved open reading frames (ORFs) (Fig. 1). No introns were found in all the genes. Two genes, trnY situated between trnA and rps10 and orf379 in F. vesiculosus were absent in T. ornata and Sargassum (Oudot-Le Secq et al. 2006), indicating that the deletion event might have occurred in the common ancestor of Turbinaria and Sargassum.

Fig. 1
figure 1

Mitochondrial genome organization of Turbinaria ornata. Genes shown on the inside of the map are transcribed in a clockwise direction, whereas those on the outside of the map are transcribed counterclockwise. Different gene types are shown as filled boxes in different colors

Full size image

The structure and organization of the T. ornata mitogenome is identical to Sargassum species. Genes are encoded on both strands of the mtDNA, suggesting the presence of at least two main transcriptional units. There is no inverted repeat in the T. ornata mitogenome. Only one tandem repeat with the period size of 22 bp was observed in the nad4 gene, which was also found in the mitogenomes of S. muticum and S. fusiforme, but not in other Fucales.

Genes in the T. ornata mitogenome

Coding sequences within the T. ornata mtDNA were initially identified by a comparison to mitochondrial gene sequences of reported Sargassum species. The 35 protein-coding genes including 17 ribosomal proteins (rps2-4, 7, 8, 10-14, and 19; rpl2, 5, 6,14, 16, and 31), 10 NADH dehydrogenase complex subunits (nad1-7, 4L, 9 and 11), three cytochrome oxidase subunits (cox1-3), three ATPase subunits (atp6, 8 and 9), apocytochrome b (cob), and secY-independent transporter protein (tatC), exist in all sequenced brown algal mitogenomes.

All protein-coding genes start with a typical initiation codon ATG in the T. ornata mitogenome, while unconventional start codons GTG, CTG and TTG are rarely used in other brown algal mtDNAs (Liu et al. 2015). In brown algal mtDNAs, most protein-coding genes terminate with TAA stop codon, which account for 63.16–89.74 %. In the T. ornata mitogenome, 26 protein-coding genes (70.27 %) terminate with TAA stop codon, seven with TAG and four with TGA (Table 2). No incomplete stop codon is found here or in other brown algal mitogenomes.

Table 2 Mitogenome features of Turbinaria ornata
Full size table

A total of 25 tRNA genes are detected in the T. ornata mitogenome, which could transport 19 amino acids, but are not sufficient for messenger RNA translation inside the mitochondria. The trnT gene responsible for the transport of threonine was neither detected in the T. ornata mitogenome, nor was it found in other brown algal mtDNAs. However, considering the amino acid Thr is present in many predicted protein products, the trnT gene might have been shifted to the nuclear genome in very early period of brown algal evolution, but such assumption needs to be determined with further genetic evidence.

Unlike the other 24 tRNA genes that could fold into the standard cloverleaf secondary structure, the tRNA gene situated between nad4 and nad5 was variable in structure and function among the different brown algal mitogenomes (Table 3). In the Ectocarpales and Laminariales a 15-bp insertion has been found in the D-loop, which has changed the conventional structure of tRNA. This modified tRNA gene might play a new role instead of old function (Oudot-Le Secq et al. 2006). The insertion was not found in the mitogenomes of Fucales, Desmarestiales, and Dictyotales. In the mtDNAs of T. ornata and Sargassum, this region is predicted as trnI with the anticodon TAT, but the anticodon is TTA for a stop codon in F. vesiculosus, and TTT for the trnK gene in D. dichotoma and Desmarestia viridis.

Table 3 The aligned sequences of tRNA genes located between nad4 and nad5 in the 21 brown algal mtDNAs, including T. ornata (To), S. muticum (Sm), S. hemiphyllum (Shm), S. fusiforme (Sf), S. horneri (Sh), F. vesiculosus (Fv), D. viridis (Dv), C. peregrina (Cp), S. lomentaria (Sl), P. fascia (Pf), E. siliculosus (Es), P. littoralis (Pl), U. pinnatifida (Up), C. costata (Cc), L. digitata (Ld), L. hyperborea (Lh), S. japonica (Saj), S. longissima (Sal), S. angustata (Saa), S. coriacea (Sac), and D. dichotoma (Dd)
Full size table

Overlapping regions

Brown algal mitogenomes usually contain 10–14 overlapping regions, with the exception of P. littoralis which contain 17. The T. ornata genes overlap by a total of 164 bp in 12 different locations from 1 to 66 bp in length (Table 2). All are found in species in the Sargassaceae. Among them, the nad1-tatC overlapping region is the largest one with the size of 66 bp, which is similar to other species in the Fucales: 60 bp in S. thunbergii, 63 bp in S. muticum, 66 bp in S. horneri and F. vesiculosus, and 90 bp in S. hemiphyllum and S. fusiforme.

Ten of the overlapping regions are highly conserved at the family level with respect to length and sequence, including the overlaps of rps8-rpl6 (4 bp), rpl6-rps2 (1), rps2-rps4 (11), rps12-rps7 (7), rps19-rpl2 (14), rps13-rps11 (11), cox2-nad4 (10), nad5-nad6 (16), nad3-rps14 (4), and rrn5-trnM (3). The rpl16-rps3 region is 17 bp in T. ornata, which is identical to that in other species of Sargassaceae except for S. fusiforme, where this region is 32 bp in size with a 15-bp insertion. Comparative analyses of overlapping regions at the order level reveal that seven of the overlapping regions (rps8-rpl6, rpl6-rps2, rps13-rps11, cox2-nad4, nad5-nad6, nad3-rps14, and rrn5-trnM) identified are highly conserved in Fucalean mitogenomes. However, only two of the overlapping regions which are exactly conserved among brown algae of the five orders are rps8-rpl6 and rpl6-rps2 (Oudot-Le Secq et al. 2006).

Non-coding sequences

The non-coding sequences comprise 1872 bp in total, approximately 5.35 % of the T. ornata mitogenome sequence. This is higher than Sargassum (4.29–4.65 %) and lower than F. vesiculosus (5.62 %) (Table 2). On the whole, the spacer content of the Fucales mitogenomes are significantly low compared with that of Desmarestiales (6.06 %), Ectocarpales (6.02–6.81 %. not including C. peregrina: 4.16 %), and Laminariales (5.83–6.49 %), indicating that the Fucales mitogenomes are more compact and economical than that of other orders in the Ectocarpales, Laminariales, Desmarestiales, and Fucales (ELDF) complex (Liu and Pang 2014c).

The total spacer size is positively correlated with brown algal genome size. We compared the spacer size with the mitogenome size of 21 brown algae and estimated their correlation coefficient to be 0.7972 (Fig. 2). Like the vast majority of the mitogenomes, the AT content of intergenic spacer regions is significantly rich in brown algal mtDNAs. In the T. ornata mtDNA, the AT content of non-coding regions is 75.48 %, which is in range of known brown algal mtDNAs from 67.41 % in Pylaiella littoralis to 83.92 % in C. peregrina.

Fig. 2
figure 2

Relationship between brown algal mitogenome size and total spacer size. Correlation coefficient is 0.7972

Full size image

Mitochondria originated from a proteobacterial ancestor through an endosymbiotic event (Liu et al. 2015). A positive correlation has been identified between the increase in AT richness and the DNA loss for these nonfunctional DNA regions in endosymbiotic bacteria (Gómez-Valero et al. 2004). Large intergenic spacer regions normally limited in number and size tend to be eliminated from mitogenomes due to deletional bias and inactivity. The largest spacer region in the T. ornata mtDNA is located between cox3 and atp6 with the size of 251 bp. We previously found that the cox3-atp6 spacer region is the most variable in Sargassum mitogenomes with respect to sequence length and identity, which was directly responsible for the variance in Sargassum genome size (Liu and Pang 2015). The size of this region is 1299 bp in F. vesiculosus containing a conserved orf379, while it has been reduced to 20–113 bp in Sargassum species of subgenus Bactrophycus.

Gene arrangements

The gene order or arrangement of mtDNAs among the species in the orders ELDF has high similarity, varying in ORF number and tRNA gene position. These ELDF arrangement patterns are apparently different from that of Dictyotales representing a more ancestral brown algal lineage (Rousseau et al. 2001; Phillips et al. 2011), in which more genes appear in different positions (Oudot-Le Secq et al. 2006; Liu and Pang 2014c).

The mitogenome sequences of Desmarestiales, Ectocarpales, Laminariales and Dictyotales are more incongruent with that of T. ornata than five other species in Fucales, considering that more gap regions are detected based on the BRIG map (Fig. 3). The overall mitogenome sequences are fairly conserved in the ELDF, and most regions of the mtDNAs have more than 70 % of sequence identity relative to T. ornata. However, several regions such as the rnl-trnK spacer, cox2 gene, cox3-atp6 spacer, rps14-rns middle region and trnP-rnl spacer, display greater inconsistency. This indicates that mitochondrial variation is more active in these regions during the process of evolution. For example, the cox2 gene in ELDF mtDNAs contains an in-frame insertion with different sizes (2.28–2.99 kb); and the rps14-rns middle region has a high frequency of gene rearrangement.

Fig. 3
figure 3

Comparison of the T. ornata mitogenome against 20 published brown algae. The innermost ring shows GC content (black), and the second innermost one shows the T. ornata mitogenome as reference sequence (purple). The remaining rings show BLAST comparisons of other complete brown algal genomes against T. ornata, which from inside to outside are Fucales (S. muticum, S. hemiphyllum, S. fusiforme, S. horneri and F. vesiculosus. aqua), Desmarestiales (D. viridis. blue), Ectocarpales (C. peregrina, S. lomentaria, P. fascia, E. siliculosus, and P. littoralis. green), Laminariales (U. pinnatifida, C. costata, L. digitata, L. hyperborea, S. japonica, S. longissima, S. angustata, and S. coriacea. yellow), and Dictyotales (D. dichotoma. red). The intensity of the ring color denotes the degree of sequence conservation at that region. Gaps or white spaces indicate highly variable regions (<50 % sequence similarity), six of which are annotated in the outermost words (black). The figure was generated using BRIG software

Full size image

Sequence identity and phylogenetic analyses

Species in orders ELDF share the identity values from 68.0 to 99.9 % (Table 4), which are significantly higher than that between D. dichotoma and the others (58.7–60.5 %). Previous studies indicated that tRNA genes were most conserved in brown algal mtDNAs, followed by rRNAs and protein-coding genes, with intergenic spacer regions being the most variable (Liu et al. 2014b, c). To examine the phylogenetic relationships within the Phaeophyceae, ML and NJ trees were constructed on the bases of three datasets: 3 rRNA genes, 25 tRNA genes, and 35 protein-coding genes. The current analyses based on different categories of genes produce some incongruent topographies, which are clearly displayed by the variation in the overall topology of phylogenetic trees (Figs. 4, 5).

Table 4 Identity matrix (%) of the 21 brown algal mtDNA sequences
Full size table
Fig. 4
figure 4

Phylogenetic trees depicting brown algal relationships inferred from a three mitochondrial rRNA genes (rnl, rns and rrn5) and b 25 mitochondrial tRNA genes. Dictyota dichotoma was designated as the outgroup. Numbers on branches indicate maximum likelihood (ML, above) and neighbor-joining (NJ, below) bootstrap values

Full size image
Fig. 5
figure 5

Phylogenetic trees depicting brown algal relationships inferred from a nucleotide and b amino acid datasets of 35 mitochondrial protein-coding genes (rps2-4, rps7, rps8, rps10-14, rps19; rpl2, rpl5, rpl6, rpl14, rpl16, rpl31; nad1-7, nad9, nad11; cob; cox1-3; atp6, atp8, atp9; and tatC). Dictyota dichotoma was designated as the outgroup. Numbers on branches indicate maximum likelihood (ML, above) and neighbor-joining (NJ, below) bootstrap values

Full size image

Based on 3 rRNAs or 35 PCG datasets, 5 resolved clades, representing 5 different orders Ectocarpales, Laminariales, Desmarestiales, Fucales, and Dictyotales, are clearly formed with strong support values. Analysis of 25 tRNAs confirmed the ELDF affinity, but failed to cluster F. vesiculosus with five species in the Sargassaceae. The inconsistency of the Desmarestiales position, and the variation of branching patterns in the interior of Ectocarpales, Laminariales, and Fucales clades are observed in the present study. Considering the common origin of brown algal mtDNAs, the observed uncertainty regarding the position and support values of some branches is closely associated with the heterogeneity in evolutionary rate of different categories of genes.